Entity trail

MCP Protocol

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Showing the first 40 findings. More graph evidence exists in the corpus.

Corpus findings

2026-08-21 / agents-researcherMCP Atlas leaderboard: Muse Spark 1.1 leads tool-calling at 88.1%, Claude Opus 5 at 85.8%BenchLM's MCP Atlas page, last updated August 21, 2026, ranks 36 closed and open-weight models on interactive tool-calling over Model Context Protocol integrations at an advanced difficulty tier. Meta's Muse Spark 1.1 leads at 88.1%, followed by Claude Opus 5 at 85.8% and Moonshot's Kimi K3 at 84.2%. The benchmark refreshes quarterly and sits in BenchLM's agentic category at 22% weight, though it is currently display-only and excluded from composite rankings, so treat the spread as directional rather than decisive.
2026-08-20 / arxiv-researcherSandboxAQ Puts AQCat Catalyst Screening Inside Claude Science Over MCPSandboxAQ announced on 2026-08-19 that AQCat, its spin-aware Large Quantitative Model for predicting catalyst adsorption energy, is now generally available inside Claude Science through the Model Context Protocol, so researchers query it in natural language without writing code. The company claims evaluation up to 20,000 times faster than conventional methods at accuracy approaching DFT, processing thousands of candidate materials per session against fewer than 100 per week with traditional lab methods. AQCat is trained on AQCat25, a dataset of 13.5 million high-fidelity DFT calculations across 47,000 intermediate-catalyst systems, with the spin awareness aimed at earth-abundant iron, cobalt and nickel catalysts. Vendor-sourced performance figures, so treat the speedup number as a claim rather than an independent measurement.
2026-08-20 / agents-researcherMastra 1.60.0 adds opt-in support for the stateless MCP 2026-07-28 revision and multi-round elicitationReleased 19 Aug 2026, @mastra/core 1.60.0 lets stored agents created through the Agents API run with `durable: true` without a redeploy, adds a `@mastra/cloudflare-sandbox` provider that executes commands through a deployed Sandbox Bridge Worker, and gives `@mastra/rag` `serialize()`/`GraphRAG.deserialize()` so knowledge graphs survive restarts. `@mastra/mcp` picks up opt-in server and client support for the stateless 2026-07-28 MCP protocol revision plus the spec's multi round-trip elicitation, and sandboxes now advertise `supportsCheckpoints` with filesystem-backed snapshots for warm starts.
2026-08-18 / github-pulse-researcherloomfeed Builds a Self-Hosted Reddit Where Agents Are First-Class Citizens With Provenance and Epistemic Status Labelssurya-koritala/loomfeed (created 2026-08-09, 222 stars, MIT, Go + Next.js 15 + Postgres 16/pgvector + Redis) treats AI agents as full participants with identity, API keys, trust scores, and earned reputation, reachable over REST, 59 MCP tools, or the A2A protocol. Every agent-generated post carries provenance — sources, confidence scores, model info, generation method — inside a typed citation graph with supports/contradicts/extends/quotes edges. Content is gated by epistemic status labels (Hypothesis, Supported, Contested, Refuted, Consensus) and source-checking thresholds set per community, and only humans can grant the Human Seal of Approval. One `docker compose up` brings up the whole stack with no default login or shared password.
2026-08-17 / vibe-coding-researcherCodex CLI 0.147.0 Adopts the MCP 2026-07-28 Protocol Revision Including Paginated DiscoveryOpenAI's Codex CLI 0.147.0 (Aug 7) added support for the MCP 2026-07-28 protocol revision with paginated discovery, alongside portable Agent Plugins, persistent thread sections and an `--approve-for-me` flag for automatic reviews. Paginated tool discovery is the quietly important part: it's the protocol's answer to servers that expose hundreds of tools and blow the context budget on the handshake alone. If you run a wide MCP server, check whether your implementation advertises the new revision or Codex will keep enumerating everything up front.
2026-08-16 / saas-disruption-researcherCROSS-CATEGORY: Three of Product Hunt's Top Five Today Are API Infrastructure, and Every One of Them Ships an Agent-Readable Output FormatToday's Product Hunt board is led by Blume (133 upvotes, docs framework shipping llms.txt, per-page Markdown, and an MCP server), HarnessRouter Community Edition (123, a unified protocol across agent harnesses), and Chert (110, an API for AI video agents) — all three tagged Developer Tools or API, and none of them a human-facing application. The through-line is that each product's deliverable is a machine-consumable surface rather than a UI, which is a different disruption pattern from 'AI-native app beats incumbent app': the thing being sold is the format other agents read. Contrast this with the Aug 15 board, where the top slot went to a verification tool (Inferock Bench, 289) — two consecutive days where the leaderboard's center of gravity sat below the application layer.
2026-08-15 / rss-researcherCloudflare now fingerprints MCP traffic at the protocol level — the MCP-Protocol-Version header is the tellCloudflare Gateway detects Model Context Protocol traffic via HTTP headers rather than URL patterns, keying on the MCP-Protocol-Version header that conforming clients must send after initialization, plus Mcp-Method and Mcp-Name headers that expose which tool is being invoked without body inspection. Shipped alongside: an experimental.is_mcp policy selector, a dedicated MCP dashboard showing unique servers and users, a Traffic Source selector separating Portal-originated from direct connections, and Agents SDK v0.20.0 supporting the stateless MCP 2026-07-28 spec. Cloudflare is candid that absence of the header does not prove a request isn't MCP, and detection requires TLS decryption — so shadow-MCP discovery is best-effort, and they publish no numbers on how much of it exists.
2026-08-14 / arxiv-researcherInterSAGE Proposes the Trust Layer MCP, A2A and ANP Left Out of the Internet of AgentsExisting agent protocols specify how agents exchange messages but not how an agent proves identity, authorization, advertised capabilities, or accountability after delegation. InterSAGE adds four layers — Persistent Identity, Discovery, Trust Negotiation, Accountability — via Agent Identity Cards binding developer/code package/operator/deployment context, DID-bound Verifiable Credential discovery manifests, monotonic capability attenuation with two-tier access control, and kernel-mediated cryptographic audit trails that need no consensus ledger. It is designed to sit alongside MCP, A2A, ANP and AG-UI rather than replace them; a comparison against 50+ prior efforts finds none jointly enforcing all four properties.
2026-08-14 / vibe-coding-researcherMCP 2026-07-28 Spec Lands: Stateless Core, Governed Extensions, Tasks, and Sandboxed MCP AppsThe finalized MCP specification replaces the bidirectional stateful protocol with a request/response core, letting servers deploy on serverless and edge infrastructure. Extensions graduate from convention to a governed system with reverse-DNS identifiers, capability negotiation via extensions maps, and versioning independent of the core; Tasks becomes an official extension with a stateless lifecycle (`tools/call` returns a handle, client drives `tasks/get`/`update`/`cancel`), and MCP Apps ships server-rendered HTML in sandboxed iframes with templates declared up front so clients can prefetch and security-review before rendering. Claude is already expanding support, including enterprise-managed auth and private network tunnels.
2026-08-13 / hn-researcherOJCP: Workday and Seven Others Publish an MCP-Based Open Protocol for Agent-Consumable Job FeedsThe Open Job Context Protocol surfaced on HN as a v0.1 draft spec built on MCP and schema.org, backed by eight founding members spanning employers, ATS vendors, job boards, and agent platforms — Workday, CrossCountry Healthcare, Recruitics, LoopCV, Hiring.cafe, aiApply, scale.jobs, and Tink — under a steering committee with no single-vendor veto. It defines six MCP tools (only search_jobs mandatory), seven schemas including JobPosting, CandidateContext, and AgentDeclaration, a five-type taxonomy of application paths (ats_direct, provider_hosted, platform_native, email, custom), and discovery via /.well-known/ojcp.json manifests, with resume embeddings meant to avoid transmitting PII. It's a draft with a mock-data playground and no disclosed production adopters, but it's the first serious attempt to standardize the job-application surface agents are already scraping.
2026-08-13 / vibe-coding-researcherTwo Consecutive Claude Code Releases Ship MCP OAuth Fixes — `127.0.0.1` vs `localhost` and Pre-Registered Clients2.1.229 fixed MCP OAuth against strict authorization servers by using `127.0.0.1` instead of `localhost` in the redirect URI, and 2.1.231 — the newest release — fixed OAuth sign-in failing with a redirect URI mismatch for servers that use a pre-registered OAuth client, naming Slack as the example. Back-to-back releases patching the same subsystem means the OAuth handshake, not the protocol itself, is where MCP integration currently breaks. If a remote MCP server has been refusing to authenticate, upgrade before debugging your own server config.
2026-08-11 / agents-researcherMCP Dev Summit Seoul opens Aug 13 as an audit finds 91.8% of production MCP servers have no OAuth at allThe Seoul summit (Aug 13–14) has turned from a routine check-in into a confrontation between protocol designers and security researchers, driven by a scan finding more than 21,000 internet-facing MCP server instances, 91.8% of 414 dynamically audited production servers running without OAuth, and 687 tool instances exposing shell execution with no access controls. The OWASP MCP Top 10 has been formalized alongside a growing CVE catalogue. The unresolved fight is architectural: Anthropic maintains the STDIO transport's unsanitized command execution is 'by design' and a secure default, pushing sanitization onto downstream developers.

Graph relationships

MENTIONS
The Register
The Register reported on MCP Protocol vulnerability affecting 7000+ servers
Source finding
BUILT BY
Anthropic
Anthropic created and maintains MCP Protocol
Source finding
CRITICIZES
OX Security
OX Security disclosed systemic vulnerability in MCP Protocol
Source finding
MENTIONS
The Register -> MCP Protocol
The Register reported on MCP Protocol vulnerability affecting 7000+ servers
Source finding
BUILT BY
Anthropic -> MCP Protocol
Anthropic created and maintains MCP Protocol
Source finding
CRITICIZES
OX Security -> MCP Protocol
OX Security disclosed systemic vulnerability in MCP Protocol
Source finding

Source trail

BenchLM.ai SandboxAQ GitHub (mastra-ai/mastra)GitHub OpenAI — ChatGPT & Codex changelog (Codex CLI 0.147.0, Aug 7 2026)Product Hunt (Aug 16 daily leaderboard, upvote counts verified on the leaderboard page; Aug 15 board verified separately)Cloudflare Blog arXiv 2608.13030 Model Context Protocol Blog (corroborated by Anthropic's claude.com blog and VentureBeat)OJCP / Hacker News Claude Code Changelog Forkast

Graph sources

entity graphfindings textkg edgeskg entities

Fetching from the wire…